On Mon, Feb 8, 2010 at 9:48 AM, Mark Kettenis <mark.kette...@xs4all.nl> wrote: > For one thing, I'd like to see some real benchmarks. Does using a > larger buffer really speed up cp? You claim moving a "head" between > reads and writes takes time, but so does moving it between reads. And > modern drives have caches that may influence the behaviour.
cp copying a 200M file without holes on a spinning laptop drive. before: 19s after: 16s Subjectively, the system generally felt a little more responsive to usual commands (cd, ls, ...), but not always. The gains are perhaps less than are worth fighting over, so I'll just let it go. Increasing the buffer to 300M (enough to fit the entire file), I could drop the copy time to 14.5s, so there's the floor for any kind of optimization along these lines.
